Multiuser Precoding Neural Network User Selection

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing wireless communication systems face challenges in effectively selecting users for multiuser precoding, especially in scenarios requiring high communication capacity, reliability, and low latency.

Innovation Solution

The proposed solution involves a device and method for selecting users for multiuser precoding using precoders in a wireless communication system. This includes transmitting configuration information related to channel state information (CSI) feedback, determining precoding vectors based on CSI feedback signals generated by a neural network model, and performing precoding for data transmission to participating devices.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If neural network model-based multiuser precoding is implemented, then communication capacity and reliability are improved, but device complexity increases due to additional processing requirements

Engineering Contradiction:

Improvecommunication reliabilityVSAvoidprocessing complexity

Core Design Contradiction:

ReliabilityVSDevice complexity

Solution Approach 1:

The patent replaces traditional mechanical signal processing systems with neural network-based intelligent processing. The encoder neural network and decoder neural network automatically learn optimal precoding patterns from channel state information, eliminating the need for complex manual signal processing algorithms and reducing overall system processing complexity while improving communication reliability.

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

Solution Approach 2:

The patent changes the operating parameters of the precoding system by using learned parameters from neural networks instead of fixed traditional parameters. The encoder neural network processes channel state information to generate encoded parameters, which are then decoded to produce precoding matrices, enabling adaptive optimization of communication performance under varying channel conditions.

Inventive Principle:

Principle #35Parameter changes

2Reliability

If iterative exclusion operations are performed for user selection, then precoding performance is improved, but processing time increases

Engineering Contradiction:

Improveprecoding performanceVSAvoidprocessing time

Core Design Contradiction:

ReliabilityVSLoss of time

Solution Approach 1:

The patent applies preliminary action by performing user selection and precoding vector determination before actual data transmission. The encoder neural network processes channel state information in advance to generate encoded parameters, and the decoder neural network pre-computes precoding matrices, eliminating the need for time-consuming iterative exclusion operations during transmission and reducing overall processing time.

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The patent substitutes traditional iterative exclusion algorithms with neural network-based user selection. The encoder and decoder neural networks automatically identify optimal users and compute precoding vectors through learned patterns, replacing manual iterative exclusion processes and significantly reducing processing time while maintaining or improving precoding performance.

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

3Measurement precision

If additional feedback information is collected for user selection, then user selection accuracy is improved, but system complexity and overhead increase

Engineering Contradiction:

Improveuser selection accuracyVSAvoidfeedback processing complexity

Core Design Contradiction:

Measurement precisionVSDevice complexity

Solution Approach 1:

The patent extracts only the essential information needed for effective user selection from the available channel state information. The encoder neural network processes and extracts relevant features from CSI, generating encoded parameters that capture the most important channel characteristics without requiring all possible feedback information, thereby reducing feedback processing complexity while maintaining user selection accuracy.

Inventive Principle:

Principle #2Taking out (Extraction)

Solution Approach 2:

The patent replaces traditional mechanical feedback processing systems with neural network-based information extraction. The encoder and decoder neural networks automatically identify and process only the critical feedback parameters needed for accurate user selection, eliminating the need for collecting and processing all possible feedback information and reducing system complexity.

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

Data Source

PatentUS20250030463A1Device and method for performing multi-user precoding in wireless communication system

Publication Date: 2025.01.23 LG ELECTRONICS INC

AI summary

The present disclosure is to perform multiuser precoding in a wireless communication system. A method of operating a device for performing multiuser precoding in a wireless communication system may comprise transmitting configuration information related to channel state information (CSI) feedback to candidate devices, transmitting reference signals corresponding to the configuration information, receiving CSI feedback signals from the candidate devices, determining precoding vectors for participating devices that are at least part of the candidate devices, performing precoding for data to the participating devices using the precoding vectors, and transmitting the precoded data. The participating devices may be determined based on information including magnitude values of precoding vectors for the candidate devices determined by a decoder neural network based on the CSI feedback signals generated by an encoder neural network.